GPL would have helped with the concerns around the distributed software (instead of just source) not clearly including attribution/copies of the license (which would also lead to a better form of notification than the conference and webpage acknowledgement). These were also the types of points Tanenbaum famously regretted regarding MINIX https://www.cs.vu.nl/~ast/intel/ despite him not having the further regrets in the article.

I do agree for the author to be _fully_ happy they would probably have wanted something even more restrictive than any traditionally "open" license like GPL, but about any choice would have better aligned with their desires than MIT.

For the copyright part, it wouldn't have lead to a different outcome. What could have been different is that Microsoft could have had difficulties in working on a GPL fork which is harder to resell (you can, but people are sometimes afraid for good or bad reasons) and so Microsoft could have proposed to the author to sell them a copy with a different license.

But reading the article, the author appears to be more disgruntled by the fact that a behemoth forked his project than the mishandling of the copyright that can be fixed with one PR (he is right to be pissed about that, but that's an easily solvable problem, I doubt Microsoft will stand against it).

I suspect that damages may also play a role in practical resolution of infringement.

There is a large difference between "they didn't put in a sentence that they needed to," and "we have 30 users who didn't get the source code that they were required to receive."

When legal reads "GPL" they go completely crazy. Had it been GPL they'd have most likely told the developers to stay really really far away from that code.